The past decade has witnessed the introduction of multiple new therapies for the treatment and supportive care of cancer patients. However, studies suggest that some cancer patients do not receive these newer, more effective treatments while others are offered these costly treatments without evidence of survival or quality-of-life benefit. While under-use has been evaluated to some degree in the setting of potentially life- saving surgery and chemotherapy, there has been very little evaluation of over-use. Our group has previously had extensive experience using the SEER-Medicare database to explore the utilization patterns of cancer treatments proven to be efficacious. We now propose to extend our studies to evaluate over-use of these newer, costly therapies. Because of the economc implications, issues concerning the appropriate use and over-use of expensive new cancer treatment and technology is of great interest. We propose to use the linked Surveillance, Epidemiology and End Results (SEER)-Medicare database to evaluate several important new costly therapies for their patterns of "appropriate use", over-use", benefits and toxicities. The treatments we will investigate have published evidence-based clinical guidelines advising appropriate use: (1) use of IV bisphosphonates (pamidronate/zoledronate) in patients with multiple myeloma or solid tumors with bone metastases;(2) use of erythrocyte stimulating agents (ESA's) in several common cancers;(3) use of the first anti-tumor antibody, rituximab (Rituxan), in patients with high and low grade lymphoma. In addition, for each of these agents we will identify the impact of patient (race, age, socioeconomic status), tumor (stage, grade), health system (teaching hospital, cancer center) and physician characteristics (specialty, age, gender) on the patterns of use, length of use and off-label use. This study will use a retrospective cohort design using the most updated SEER-Medicare and the American Medical Association Master file. In order to improve performance outcomes and quality of care while controlling health care costs, we need to understand patterns of care so that targeted interventions to improve quality can be implemented. The information obtained will help guide policy makers to improve the reglation of the use of the increasing cadre of treatment options for cancer treatments. PERFORMANCE SITE(S) Columbia University Medical Center, NY, NY. The past decade has witnessed the introduction of multiple new therapies for the treatment and supportive care of cancer patients. However, studies suggest that some cancer patients do not receive these newer, effective treatments while others are offered these costly treatments without evidence of benefit. We propose to use the linked Surveillance, Epidemiology and End Results (SEER)-Medicare database to evaluate several important new costly therapies for their patterns of "appropriate use", over-use", benefits and toxicities. PUBLIC HEALTH RELEVANCE: The past decade has witnessed the introduction of multiple new therapies for the treatment and supportive care of cancer patients. However, studies suggest that some cancer patients do not receive these newer, effective treatments while others are offered these costly treatments without evidence of benefit. We propose to use the linked Surveillance, Epidemiology and End Results (SEER)-Medicare database to evaluate several important new costly therapies for their patterns of "appropriate use", "over-use", benefits and toxicities